Models
131 models from 46 makers — agent reasoning models, independent API resources, and image, video and audio models. Choose Use for the composer or Details for API access.
https://gateway.agentsky.dev/v1curl https://gateway.agentsky.dev/v1/messages \ --header "Authorization: Bearer $SKY_API_TOKEN" \ --header "anthropic-version: 2023-06-01" \ --header "Content-Type: application/json" \ --data '{"model": "anthropic/claude-haiku-5-5", "max_tokens": 256, "messages": [{"role": "user", "content": "Hello"}]}'
Fast adaptive reasoning for high-volume coding and assistant workloads.
Adaptive reasoning for everyday coding and agentic work.
Anthropic's flagship model for complex agentic coding and long-horizon professional work.
OpenAI's most efficient GPT-6 tier for focused, high-volume workloads.
xAI's latest flagship for coding, agentic tasks, and knowledge work, with four reasoning-effort levels and a 500K-token context.
OpenAI's flagship for the hardest end-to-end work across complex reasoning, coding, computer use, research, and document creation.
The most intelligent Flash model, engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows, taking text, images, audio, video, and documents.
A natively multimodal Flash model for efficient coding and long-horizon agent tasks, taking text, images, and video.
The current GLM flagship for software engineering and agent workloads, with configurable reasoning effort.
The higher-capability V4 text model, with thinking mode on by default and a million-token window.
Moonshot's flagship multimodal model, with always-on reasoning and a selectable effort level.
A coding-oriented multimodal model supporting extended reasoning.
The higher-throughput serving variant of K2.7 Code, at twice the price.
xAI's lower-cost coding model for agentic software engineering and workflow tasks. This API model is separate from the Grok Build agent harness.
A vision-language model taking images, video, and documents alongside text.
The lightweight, lower-priced member of the GLM-4.5 family.
Aion 3.5 is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models.
Aion 3.5 Mini is a multi-model roleplaying and storytelling system from AionLabs, built on the GLM family of models.
Beam 1.0 is Reflection's open-weight reasoning model for coding and long-horizon agentic workflows.
Claude Fable 5.2 is Anthropic's most capable model for the most demanding technical work, with gains over Fable 5.1 in coding, reasoning, and vision.